Thermal oxidation to regenerate sulfone poisoned Pd-based catalyst: Effect of the valence of sulfur

Article abstract of DOI:10.1039/c4ra03546a

Sulfur deactivation is a serious problem which largely limits the industrial application of noble metals as catalysts. Here we report a thermal oxidation method to regenerate sulfone poisoned Pd/C catalyst applied in the hydrogenation of sodium-m-nitrobenzene sulfonate (SNS). It was found that the initial activity of Pd/C catalyst could be substantially recovered after treating it in air at temperatures as low as 100 °C. And the catalyst could be reused for at least 20 times without the significant loss of activity. The properties of deactivated and regenerated catalysts were studied in detail by BET measurement, X-ray photoelectron spectroscopy (XPS), temperature programmed desorption (TPD), and Fourier transform infrared spectroscopy (FT-IR). The results indicated that the main surface sulfur species found on deactivated and regenerated Pd surfaces were S_n and sulfate (SO₄), respectively. The change of the valence of sulfur species was found to be the key factor influencing the catalytic activity of the Pd-based catalyst. the Partner Organisations 2014.

Full text of DOI:10.1039/c4ra03546a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Thermal oxidation to regenerate sulfone poisoned
Pd-based catalyst: effect of the valence of sulfur
Cite this: RSC Adv., 2014, 4, 33347
Tieyong Xu, Qunfeng Zhang, Dahao Jiang, Qiuxia Liang, Chunshan Lu, Jie Cen
*
and Xiaonian Li
Sulfur deactivation is a serious problem which largely limits the industrial application of noble metals as
catalysts. Here we report a thermal oxidation method to regenerate sulfone poisoned Pd/C catalyst
applied in the hydrogenation of sodium-m-nitrobenzene sulfonate (SNS). It was found that the initial
activity of Pd/C catalyst could be substantially recovered after treating it in air at temperatures as low as
100 ꢀC. And the catalyst could be reused for at least 20 times without the significant loss of activity. The
properties of deactivated and regenerated catalysts were studied in detail by BET measurement, X-ray
photoelectron spectroscopy (XPS), temperature programmed desorption (TPD), and Fourier transform
infrared spectroscopy (FT-IR). The results indicated that the main surface sulfur species found on
deactivated and regenerated Pd surfaces were Sn and sulfate (SO4), respectively. The change of the
valence of sulfur species was found to be the key factor influencing the catalytic activity of the Pd-based
catalyst.
Received 18th April 2014
Accepted 3rd July 2014
DOI: 10.1039/c4ra03546a
www.rsc.org/advances
energy consuming.12 The second approach is to improve the
sulfur tolerance of metal catalysts. There are typically two
1. Introduction
approaches employed by researchers: (a) changing the acid-base
properties of the carrier; (b) changing the electronic and
structural properties of the metal particles by adding a second
metal. Concerning the former, it was reported that the sulfur
tolerance of metal catalysts could be greatly enhanced by using
acidic supports such as zeolites.13,14 While the resistance of
bimetallic catalysts to sulfur poisoning is usually the result of
decreasing the affinity of sulfur to metals through electronic
effects or structural changes.15–17 Some researchers also
proposed a sulfur and hydrogen intra-particle spillover mech-
anism to explain the improved sulfur tolerance of bimetallic
catalysts.18 However, the activity of bimetallic catalysts in a
sulfur-containing environment was still substantially lower
than those operating under sulfur-free conditions.15 Another
important way to enhance sulfur tolerance is to use metal
suldes as catalysts, such as PdS4 (ref. 19) and Rh2S3.20 Though
high stability of these catalysts can be achieved, their catalytic
activity is very low, mostly due to the harsh reaction conditions,
e.g. relatively high temperatures and pressures, making the
process energy consuming and uneconomic.19 Till now, it is still
a big challenge to overcome the problem of sulfur poisoning.
Therefore, there is an urgent requirement to nd an easy and
efficient method to regenerate sulfur-poisoned metal catalysts.
Pd-based catalysts have been proven to be the promising
metal catalysts for their excellent catalytic performance in the
hydrogenation process of nitrobenzene sulfonic acid,12 but the
information about the regeneration of sulfur-poisoned Pd
catalysts is scarcer. The regeneration by thermal treatment with
Aminobenzene sulfonic acids, such as m-aminobenze-
nesulfonic acid (m-ABSA), 1-amino-8-naphthol-3,6-disulfonic
acid (H-acid), and 6-chloro-3-aminotoluene-4-sulfonic acid are
important dye and medicine intermediates.1–3 They are usually
prepared by the reduction of the corresponding nitro-
compounds, using iron powder, hydrogen, etc. as reducing
agents.4 In contrast to the conventional iron-acid reduction
process, the catalytic hydrogenation method using metal cata-
lysts, such as nickel-, platinum-, and palladium-based catalysts
is more environmentally friendly and can produce much higher
quality products.5 However, these catalysts suffer from the
problem of sulfur poisoning caused by sulfone, which is the
inevitable byproduct of the sulfonation reaction in the
production process (Scheme 1).6
Industrial catalysts usually operate under complex condi-
tions and suffer from deactivation caused by sulfur poisoning,
carbon deposition, sintering, the blocking of the pore structure
by organic polymers, etc.7–11 Among which, sulfur poisoning is
considered as one of the major challenges in the synthesis of an
active and stable catalyst.9–11 Removing sulfur-containing
impurities in industrial reactants is one way to treat the sulfur
poisoning problem and can help maintain high activity of
industrial catalysts; however, this process is usually highly
Industrial Catalysis Institute of Zhejiang University of Technology, State Key
Laboratory Breeding Base of Green Chemistry Synthesis Technology, Hangzhou,
310014, P. R. China. E-mail: xnli@zjut.edu.cn; Fax: +86-571-88320667; Tel: +86-
571-88320002
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 33347–33354 | 33347

View Article Online
RSC Advances
Paper
Ltd.) support (the ratio of support to water was 1 g/10 mL) to
obtain a Pd nominal loading of 5 wt%. The slurry was vigorously
ꢀ
stirred at 80 C for 5 h. The pH of the slurry solution was then
adjusted to 8–9 by dropwise addition of NaOH aqueous solution
(10%). Eventually, the precipitated PdO$H2O supported on
carbon was reduced by hydrazine hydrate, ltered, and dried
under vacuum at 383 K overnight. The preparation of palladium
nanoparticles (NPs) was done following a similar procedure.
Scheme 1 Equations of synthesis of m-aminobenzenesulfonic acid
and its following hydrogenation process.
2.2. Deactivation of Pd/C and Pd NPs by 3-nitrophenyl
sulfone and their regeneration
hydrogen was reported to be useful to recover the activity of the
catalysts, however the temperature used was relatively high
(>400 ꢀC).21–24 Liquid oxidation using hydrogen peroxide (H2O2)
or sodium hypochlorite (NaClO) as oxidants is an effective way
to remove sulfur species from metal surface and has been
successfully employed to recover sulfur deactivated Pd-based
catalyst, in spite that the major disadvantage of this regenera-
tion process is the production of waste water.25,26 In addition,
both the deactivation of Pd-based catalysts by sulfur and the
corresponding regeneration processes have not previously been
investigated in details. In comparison, thermal oxidation using
air as oxidant is an ideal choice in terms of environment and
economics, and the successful application of it in regenerating
nickel catalyst has proven its potential in the regeneration of
sulfur-poisoned metal catalysts.7,27–30 As for palladium, only a
few literature can be found on the oxidation of sulfur atoms
adsorbed on Pd surface and most of the researches still stay in
the theoretical stage.31,32 Therefore, it seems adequate to try to
apply thermal oxidation to regenerate sulfur poisoned Pd-based
catalysts.
The pretreatment of Pd NPs or Pd/C catalyst by 3-nitrophenyl
sulfone was carried out in a 500 mL high-pressure stainless steel
autoclave. Typically, 2.0 g of 5% Pd/C, 1.0 g of 3-nitrophenyl
sulfone (Hebei Jianxin Chemical Co. Ltd.), or 0.1 g of Pd NPs,
1.0 g of sulfone (Aladdin Chemistry Co. Ltd.) and 150 mL of
deionized water were introduced into the autoclave. Then the
autoclave was sealed and purged with the reaction gas (N2 or H2
for Pd/C catalyst and H2 for Pd NPs) for three times to remove
the air inside. The reaction was carried on at 80 ꢀC and 0.8 MPa
with a stirring rate of 1100 rpm for 3 h. The Pd/C catalyst (or Pd
NPs) was ltered and washed thoroughly with ethanol and
deionized water, respectively. The deactivated Pd/C catalyst and
Pd NPs were denoted as Pd/C–X (X ¼ N2 or H2) and Pd NPs-H2,
respectively, where X was the reaction gas.
The regeneration experiment of Pd/C catalyst was performed
by thermal treatment of the poisoned Pd/C–H2 sample at 100 ꢀC
in different gas atmospheres for 8 h. The sample was denoted as
Pd/C–H2–Y (Y ¼ air, Ar, H2), where Y was the regeneration gas.
The Pd/C catalyst deactivated during the hydrogenation of
ꢀ
industrial SNS was treated only in air at 100 C. The Pd/C–H2
In this paper, we placed our interests in recovering the
activity of sulfur poisoned Pd-based catalysts by thermal
oxidative treatment in air and investigate the correlation
between the states of sulfur species on Pd surface and the
activity of Pd-based catalysts in details. We initially studied the
poisoning process of 3-nitrophenyl sulfone to Pd/C catalyst for
the hydrogenation of sodium-m-nitrobenzene sulfonate (SNS).
The properties of deactivated Pd/C catalysts were then analysed
by BET measurement, temperature programmed desorption
(TPD) to nd out the reasons for catalyst deactivation. The
possible reasons for the Pd/C catalyst regenerated by thermal
oxidation treatment were also discussed. Based on the results, a
thermal oxidation method developed appeared to be very easy
and efficient to recover sulfone-poisoned Pd/C catalyst and
might be highly applicable in the regeneration of other sulfur-
poisoned noble metals.
catalyst thermally treated in air at different temperatures were
denoted as Pd/C–H2–air–Z, where Z was the temperature (ꢀC)
used.
2.3. Catalytic tests
The SNS hydrogenation reaction was carried out in a 500 mL
stainless steel autoclave. Typically, 0.2 g of 5% Pd/C catalyst,
20.0 g of SNS and 100 mL of deionized water were fed into the
reactor. The autoclave was sealed and purged by H2 gas for three
times. The reaction was performed under the hydrogen pres-
sure of 0.6–0.8 MPa at 80 ꢀC with the stirring rate at 1100 rpm.
Aer the reaction, the Pd/C catalyst was ltered and washed
with deionized water several times. The products were analyzed
by liquid chromatography (Agilent 1100 series) equipped with a
UV detector (wavelength, 230 nm), and 0.5% (NH4)2CO3
(pH ¼ 6) solution was used as the mobile phase.
The hydrogenation of sulfone was carried out in a 500 mL
stainless steel autoclave. In a typical experiment, the 5%Pd/C
catalyst, sulfone, and solvent were fed into the reactor. The
reaction conditions were the same as that of SNS hydrogena-
2. Experimental
2.1. Preparation of Pd/C catalyst and palladium
nanoparticles
Pd/C catalyst was prepared by adding a desired volume of tion. And the liquid products were analyzed by GC-MS (Agilent
H2PdCl4 (Hangzhou Kaida Metal Catalyst & Compounds Co. 6890GC-5973 MS equipped with a 25 m-HP-5 capillary column,
Ltd.) aqueous solution (0.05 gmetal mLꢁ1) into an aqueous slurry MS detector). The sulfur containing compounds both in liquid
of activated carbon (C, Shaowu Xinsen Chemical Industry Co. and gas products were also monitored by online GC (Fu Li 9790
33348 | RSC Adv., 2014, 4, 33347–33354
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
GC equipped with a 30 m-FFAP capillary column, FPD detector, reasonable to assume that 3-nitrophenyl sulfone is the main
detection limit: 0.1 ppm).
cause behind the severe deactivation of Pd/C catalyst in indus-
trial application.
Some studies pointed out that 3-nitrophenyl sulfone itself
could not poison metal catalysts extensively as the sulfur atom
was believed to be totally shielded by oxygen.26,33 Therefore, we
suspect that there might be certain reactions happened between
3-nitrophenyl sulfone and hydrogen over Pd/C catalyst. To verify
our hypothesis, Pd/C catalyst was pretreated with 3-nitrophenyl
sulfone under N2 (Pd/C–N2) and H2 (Pd/C–H2) as comparison. In
Fig. 2, we can see that the initial activity of Pd/C catalyst was
only inuenced signicantly by the pretreatment in H2 (ꢂ14.5%
of fresh Pd/C), also suggesting that the adsorbed 3-nitrophenyl
sulfone itself was not enough to poison the catalyst. Therefore it
was likely that a new sulfur species was produced by the reac-
tion between 3-nitrophenyl sulfone and H2 over Pd/C catalyst.
To further investigate the transformation of sulfone, the
hydrogenation of sulfone over Pd/C catalyst was performed
(Table 1). Aniline, cyclohexylamine for 3-nitrophenyl sulfone
(entry 1 and 2) and benzene, cyclohexane for sulfone (entry 3
and 4) products were observed, indicating the hydrogenolysis of
C–S bond over Pd/C catalyst. Due to the strong conjugation
between S]O and C]C of benzene ring in sulfone, the direct
cleavage of C–S bond in sulfone was considered to be diffi-
cult.33–35 Thus to better understand the hydrogenolysis process,
the sulfur-containing products both in liquid and gas phases
were further analyzed for entry 4 as a representative by GC-FPD
(ame photometric detector). In the liquid phase, the only
detected sulfur species was cyclohexylsulfonylbenzene (Scheme
2) and its structure was conrmed by GC-MS. The results indi-
cated that the hydrogenolysis process of sulfone over Pd/C
catalyst may be as follows: (1) the hydrogenation of one of the
benzene ring in sulfone to form cyclohexylsulfonylbenzene,
therefore the conjugative structure between S]O and C]C of
2.4. Catalyst characterizations
The BET surface area and porous parameters of the Pd/C cata-
lysts were determined by nitrogen physical adsorption–
desorption at ꢁ196 ꢀC. In a typical experiment, a 50 mg sample
was heated to 180 ꢀC and held for 12 h to remove the adsorbed
species. Nitrogen adsorption isotherm was measured using a
NOVA 1000e surface area analyzer (Quantachrome Instruments
Corp.). X-ray diffraction (XRD) measurements of the Pd/C
catalysts were performed using an X'Pert diffractometer (PNA-
lytical Co.) equipped with a Cu K radiation source that was
operated at 60 kV and 55 mA. XPS spectroscopy analyses were
performed with a Kratos AXIS Ultra DLD instrument. Al Ka
radiation at an energy scale calibrated versus adventitious
carbon (C1s peak at 284.6 eV) was used. The surface composi-
tions of the samples were determined from the peak areas of the
corresponding origin tting lines.
In order to eliminate the interference of carbon support,
palladium nanoparticles (Pd NPs) were prepared and used for
Fourier transform infrared (FT-IR) analysis. The FT-IR spec-
troscopy of Pd NPs was recorded by a Necolet 6700 spectrom-
eter. The Pd NPs–H2 sample was diluted in KBr pressed into
pellets, and pressed into tablets. The regeneration of Pd NPs–H2
ꢀ
was performed by heating the pellets at 100 C under air for 8,
12 h, respectively. All FT-IR spectras were recorded with 64
scans at a resolution of 4 cmꢁ1
.
The TPD experiment with sulfone deactivated Pd/C catalyst
was carried out in a ow setup operating at atmospheric pres-
sure in helium. The gas composition leaving the reactor was
monitored by an online mass spectrometer (Omnistar). In a
typical measurement, 100 mg of Pd/C catalyst was loaded into a
tubular quartz reactor with aꢀn inner diameter of 20 mm. The
catalyst was rst dried at 80 C for 30 min in owing helium.
Subsequently, the catalyst was heated in helium at a ow rate of
30 mL minꢁ1 with a heating ramp of 10ꢀC minꢁ1, and the
products were monitored by the online MS.
3. Results and discussion
3.1. Transformation of 3-nitrophenyl sulfone and Pd/C
deactivation
3-Nitrophenyl sulfone is an inevitable byproduct during the
synthesis of m-nitrobenzene sulfonic acid (m-NBSA) and it is the
main sulfur-impurity remaining in industrial reactant before
the hydrogenation process. In order to investigate its effect on
the activity of Pd/C catalyst, chemically pure grade of reactant
(CP, 99%) was employed to compare with industrial reactant
(IN), the results are shown in Fig. 1. For analytical reactant (CP,
99%), the average reaction rates of Pd/C catalꢁys1t over three
continuous runs were 7.40, 6.73, 5.70 mmol gPd sꢁ1, respec-
tively, indicating good performance. Upon the addition of
3-nitrophenyl sulfone (CP + Sulfone), the average reaction
rate of Pd/C catalyst decreased dramatically similar to the
Fig. 1 Effect of 3-nitrophenyl sulfone on the activity of Pd/C catalyst
used for the hydrogenation of SNS. CP: chemically pure grade of
reactant (99%), IN: industrial reactant, CP + sulfone: chemically pure
grade of reactant (99%) with 0.5% mass ratio of 3-nitrophenyl sulfone
(reaction conditions: 0.2 g cat., 20.0 g SNS, 100 mL H2O, 80 ꢀC, 0.6–
behavior of the catalysts with industrial reactant. So that it is 0.8 MPa H2, 1100 rpm).
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 33347–33354 | 33349

View Article Online
RSC Advances
Paper
pore structure by organic polymers, and/or the coverage of
active sites by strong adsorbents, such as sulfur.7,8,36 To get clear
insight into the deactivation process of Pd/C catalyst, the
properties of fresh and poisoned catalysts were analyzed
through BET, XRD, and He-TPD.
3.2.1 BET. As shown in Table 2, both the surface areas and
the corresponding pore volumes of CP-Pd/C and Sulfone-Pd/C
catalysts illustrated an obvious decrease trend compared with
fresh Pd/C, indicating that organic substances, such as the
products, were accumulated in the pores of carbon support
once used. This might be the cause of the decrease in the
activity of the fresh catalyst with subsequent recycles (Fig. 1).
But the physical properties (surface area and pore volume) of
the spent catalysts (CP–Pd/C and sulfone–Pd/C) were similar.
In fact, CP–Pd/C catalyst showed good activity and excellent
recycling performance (as shown in Fig. 1) for the hydroge-
nation of SNS. Therefore, the deactivation of Pd/C catalyst
caused by 3-nitrophenyl sulfone could not be simply ascribed
to the loss of surface area and pore volume.
Fig. 2 Conversion as a function of reaction time of SNS hydrogenation
over Pd/C catalysts pretreated with 3-nitrophenyl sulfone in N2 and H2
(reaction conditions: 0.2 g cat., 20.0 g chemically pure grade of SNS
(99%), 100 mL H2O, 80 ꢀC, 0.6–0.8 MPa H2, 1100 rpm).
3.2.2 XRD. The Pd particle size of fresh and sulfone deac-
tivated Pd/C catalysts were characterized by XRD to determine
the change in the structure and the growth of crystallite. Fig. 3
also showed XRD patterns of fresh and sulfone deactivated Pd/C
catalysts. Three diffraction peaks at ꢂ39.9, 46.7, and 68.2ꢀ
correspond to the (111), (200), and (220) planes of face-centered
cubic (fcc) structure crystalline Pd, respectively.37 It should be
noted that no signicant reection shi in XRD was observed
for the sulfone deactivated Pd/C catalyst, which might be
attributed to the relatively low amount of unsaturated sulfur
species.18 The Pd particle size determined by XRD for fresh and
sulfone Pd/C catalyst was 14.4 and 14.2 nm, indicating no
signicant sintering of palladium crystallites for the sulfone
deactivated catalyst.
3.2.3 He-TPD. We have pointed out in 3.1 that the hydro-
genolysis of sulfone would generate sulfur-containing
compounds. In order to understand the interaction between
sulfur fragments and palladium surface, the sulfone deacti-
vated Pd/C catalyst was characterized by He-TPD. As exemplied
in Fig. 4, C6H6, associated to the relative mass m/z ¼ 78, and
SO2, corresponding to m/z ¼ 64, were detected at exactly the
same temperature (ꢂ330 ꢀC). The results indicated that the
generation of C6H6 and SO2 might be attributed to the thermal
decomposition of benzene sulnic acid. In addition, peaks of
thiophenol (m/z ¼ 110), hydrogen sulde (m/z ¼ 34) and
cyclohexane (m/z ¼ 84) and cyclohexanethiol (m/z ¼ 116) were
also detected in TPD experiment. The results further conrmed
Table 1 Liquid products distribution of the hydrogenation of sulfone
over Pd/C catalysta
Sulfone
Products distribution (%)
Substrate
Cat./sub.
Entry R1
R2
(g gꢁ1
)
Others
(1)
(2)
(3)
(4)
NO2 NH2
NO2 NH2 10
0.2
93.3
78.5
98.1
9.9
0.9
0.3
0.1
0.1
0.4
–
5.7
20.8
1.8
H
H
H
H
0.2
20
10.7
2.1
77.3
a
Reactions conditions: 80 ꢀC, 120 mL THF, 0.8 MPa hydrogen with (1)
0.2 g cat., 1.0 g substrate, 3 h; (2) 2.0 g cat.,0.2 g substrate, 10.34 h; (3) 0.2
g cat., 1.0 g substrate, 6 h; and (4) 2.0 g cat., 0.1 g substrate, 6 h.
benzene ring was destroyed; (2) the cleavage of C–S bond in
cyclohexylsulfonylbenzene to form benzene and cyclohexane. In
the gas phase, no sulfur species (for example H2S, SO2) was
detected even though FPD detector with very low detection limit
(0.1 ppm) was used. This nding implied that only one C–S
bond in sulfone was broken over Pd/C catalyst to form sulfur
species.
3.2 Analysis of physiochemical properties of deactivated
Pd/C catalyst. Generally speaking, the catalyst deactivation is
oen caused by the sintering of metal particles, the blocking of
Table 2 Physicochemical properties of Pd/C catalyst
Sample
SBET (m2 gꢁ1
)
Vpore (cm3 gꢁ1
)
Pdc (wt%) dPdd (nm)
Fresh-Pd/C
CP-Pd/Ca
1936
1277
1.217
0.843
0.846
4.7
4.5
4.6
14.4
14.5
14.2
Sulfone-Pd/Cb 1297
a
Spent catalyst for CP reactant. b Spent catalyst for CP reactant with 3-
nitrophenyl sulfone. c Determined by ICP-MS. Determined by XRD.
d
Scheme 2 Formula of cyclohexylsulfonylbenzen
33350 | RSC Adv., 2014, 4, 33347–33354
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
sulfur species from Pd surface. According to literatures, the
poisoning ability to metal catalysts can be largely weakened
once sulfur atom occupied by oxygen.26,36,38
With these considerations, a novel thermal oxidation
method was adopted to recover the activity of sulfone-poisoned
Pd/C catalyst (see Fig. 5). The initial activity of 3-nitrophenyl
sulfone deactivated Pd/C catalyst (Pd/C–H2) was recovered
completely aer thermally treated in air at 100 ꢀC. In addition,
the reaction time was largely shortened from 413 min (Pd/C–H2)
to 105 min (Pd/C–H2–air), indicating that the thermal oxidation
method was effective to regenerate 3-nitrophenyl sulfone
deactivated Pd/C catalyst.
3.3.1 Role of temperature in recovering sulfone-poisoned
Pd/C. As for sulfur deactivated Ni-based catalyst, the tempera-
ture required for regꢀeneration by controllable oxidation was
relatively high (ꢂ800 C), which was related to the high activa-
tion energy of the sulfur oxidation.7 The activation energy of the
sulfur oxidation on Pd is much lower than that on Ni.32,38 Got-
terbarm et al. found out that sulfur atom could be removed
from Pd surface between 127 and 227 ꢀC.32 Therefore, it is
reasonable that thermal oxidation to recover the 3-nitrophenyl
sulfone deactivated Pd/C catalyst can be operated at low
temperatures.
In this regard, low temperatures (60–120 ꢀC) were
employed to treat 3-nitrophenyl sulfone deactivated Pd/C
catalyst (Pd/C–H2) in air and their effect on the recovered
catalyst activity was shown in Fig. 6. When the temperature
adopted was as low as 60 ꢀC (Pd/C–H2–air-60), no recovery
sign could be seen on the activity of catalyst as compared to
that of untreated poisoned catalyst (Pd/C–H2, see Fig. 5).
However, the activity of Pd/C–H2 catalyst could be signi-
cantly increased as the temperature increased to 80 ꢀC and
further to 100 ꢀC. When the temperature was >100 ꢀC, the
initial activity of Pd/C–H2 was substantially recovered. And
further increasing the treated temperature to 120 ꢀC would
not dramatically inuence the activity. Therefore, the
temperature as low as 100 ꢀC seems to be enough for the
regeneration of sulfone-poisoned Pd/C catalyst.
Fig. 3 XRD patterns of (a) fresh, (b) sulfone deactivated Pd/C, (c)
regenerated Pd/C catalysts. The regeneration was conducted by
thermally treated in air at 100 ꢀC for 8 h.
that sulfone was reduced on palladium surface by the cleavage
of C–S bond. The cleavage of C–S bond of sulfone resulted in the
formation of benzene sulnic acid, and the further hydroge-
nation process would form thiophenol and hydrogen sulde
etc., as shown in Scheme 3.
With these information, the nature of Pd/C catalyst deacti-
vated by sulfone may be concluded as follows: the active sites
were blocked by coordinatively unsaturated sulfur species (such
as benzene sulnic acid, thiophenol and hydrogen sulde by
cleavage of C–S bond in sulfone), which were strongly chem-
isorbed on the metal surface, resulting in the deactivation of the
Pd/C catalyst.
3.3 Pd/C catalyst regenerated by thermal oxidation
Based on the analysis above, we have claried that the strong
adsorption of sulfur species on Pd surface is the main reason
that causes the deactivation of Pd/C catalyst. Therefore the key
to regenerate sulfone-poisoned Pd/C catalyst is the removal of
Fig. 4 TPD profiles of Sulfone-Pd/C catalyst.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 33347–33354 | 33351

View Article Online
RSC Advances
Paper
Scheme 3 Possible pathways for sulfone transformation over Pd/C catalyst.
3.3.2 Role of air in recovering sulfone-poisoned Pd/C. As
reported, thermal treatment in inert atmosphere or reductive
atmosphere have been employed to recover the activity of
sulfur-deactivated metal catalysts.27,28,39 In our work, Ar and H2
were compared with oxidative atmosphere (air) to study the
inuence of treatment atmospheres on the regeneration of
deactivated Pd/C catalyst (Pd/C–H2) (Fig. 7). It can be found that
Pd/C–H2 catalyst thermally treated in air could substantially
recover its initial activity. The activity of catalyst thermally
treated in Ar (Pd/C–H2–Ar), nevertheless, was almost identical to
that of untreated sample (see Fig. 5, Pd/C–H2). This is in
agreement with previous study by Monzon et al.39 who found a
´
similar behavior for H2S poisoned Pd surfaces. It was likely due
to the strong bonding between sulfur and metallic sites,
thermal treatment without the aid of oxygen was not sufficient
to remove sulfur species from metal surface for the reactivation
to happen.
Fig.
5 The regeneration performance of 3-nitrophenyl sulfone
deactivated Pd/C catalyst (Pd/C–H2) by thermally treated in air at
100 ꢀC for 8 h (Pd/C–H2–air) (reaction conditions identical to those
shown in Fig. 2).
Pd/C catalyst regenerated under H2 atmosphere (Pd/C–H2–H2)
showed even lower activity. It was reported that the
Fig. 7 Effect of treated atmospheres on the performance of 3-nitro-
phenyl sulfone deactivated Pd/C catalyst (Pd/C–H2) for the hydroge-
nation of SNS (the Pd/C–H2 catalyst was thermally treated in air, Ar,
and hydrogen at 100 ꢀC for 8 h, reaction conditions identical to those
shown in Fig. 2).
Fig. 6 Effect of treated temperature on the performance of 3-nitro-
phenyl sulfone deactivated Pd/C catalyst (Pd/C–H2) for the hydroge-
nation of SNS (the Pd/C–H2 catalyst was thermally treated in air at 60–
120 ꢀC for 8 h, reaction conditions identical to those shown in Fig. 2).
33352 | RSC Adv., 2014, 4, 33347–33354
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
temperatures required for the regeneration of sulfur poisoned
Pd-based catalysts in H2 was >400 ꢀC.21,22 The H2–TPR results
published in our recent paper conrmed that sulfur species on
Pd were reduced by hydrogen and released as the form of H2S
between ꢂ327 and 450 ꢀC.19 The temperature we used here was
only 100 ꢀC, so it was difficult to remove sulfur species from
metal surface. Instead, the new sulfur species formed in the
reduction of 3-nitrophenyl sulfone by H2 in low temperatures
could further deteriorate Pd/C catalysts as discussed in
Section 3.1.
The thermal treatment method of regenerating of sulfone-
poisoned Pd/C using air as reaction atmosphere has consid-
erable advantages over H2 atmosphere. Firstly, the supply
of air is abundant, making this method more economic.
Secondly, low temperature adopted can squeeze energy
usage and avoid the sintering of metal particles more
effectively.
Fig. 9 FT-IR spectras of sulfone deactivated Pd NPs (1) thermally
treated at 100 ꢀC in Ar for 8 h and Pd NPs thermally treated at 100 ꢀC in
air for (2) 8 h, and (3) 12 h.
In order to study the role of air in the regeneration process,
the deactivated and regenerated Pd/C catalysts were further
studied by XPS (Fig. 8) and FT-IR (Fig. 9). As no water vapor was
allowed in XPS and FT-IR measurement, the deactivated sample
dried in Ar was taken as the best representative of the deacti-
vated Pd/C catalyst.
The Pd 3d5/2 XPS peaks of deactivated catalyst thermally
treated in Ar (Pd/C–H2–Ar) could be deconvoluted into two
peaks, namely, metallic palladium (BE at 335.4 eV) and palla-
dium bonded with Sn & S6+ (BE at 336.7 eV).19,40 But the Pd 3d5/2
XPS peaks of Pd/C–H2–air appeared to be quite different: a new
peak at 336.1 eV which could be assigned to oxidic palladium40
had replaced the peak of PdSx. It means the sulfur species on
palladium surface was occupied by oxygen aer thermally
treated in air. The removal of sulfur species from palladium
surface is accounted for the almost complete recovery of initial
activity of Pd/C catalyst.
stretching in coordinative saturated groups, such as SO3H
groups.41 The appearance of SO2 stretching clearly indicates
that the sulfur atoms accessible were coordinated by oxygen,
which largely weakened the interaction between Pd and
sulfur.
3.3.3 Recycling of regenerated Pd/C catalyst. The stability
of Pd/C catalyst regenerated by thermal oxidation was also
tested for the hydrogenation of SNS. Aer each run, the
catalyst was cleaned by deionized water thoroughly and then
treated in air at 100 ꢀC for 4 h to recover its activity before
reused with fresh reaction mixture in a subsequent run. As
shown in Fig. 10, the activity of Pd/C catalyst slightly dropped
in the rst few runs and remained quite steady aerwards.
The regenerated Pd/C catalyst were successfully recycled for at
least 20 times without signicant loss of the activity and kept
100% of SNS conversion.
As shown in Fig. 9, no sulfur related group was visible for
Pd NPs treated in Ar, which might be caused by the relatively
weak vibrational signal of sulfur containing groups. A strong
band of ꢂ1375 cmꢁ1 was clearly visible aer thermal treat-
ment of Pd NPs in air, corresponding to the asymmetric SO2
Fig. 10 Recycling performance of regenerated Pd/C catalyst used for
the hydrogenation of SNS. Reaction conditions: 1.31 g 5%Pd/C, 131 g
Fig. 8 XPS spectra of Pd/C–H2 catalysts regenerated under different industrial SNS, 200 g H2O, 90 ꢀC, 0.8 MPa H2, 1100 rpm. The Pd/C
atmospheres. The temperature chosen for treatment was 100 ꢀC.
catalyst was thermally treated in air at 100 ꢀC for 4 h before next cycle.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 33347–33354 | 33353

View Article Online
RSC Advances
Paper
16 F. Menegazzo, P. Canton, F. Pinna and N. Pernicone, Catal.
Commun., 2008, 9, 2353.
4. Conclusions
17 J. A. Rodriguez and J. Hrbek, Acc. Chem. Res., 1999, 32, 719.
18 J. N. G. Stanley, K. Worthington, F. Heinroth, A. F. Masters
and T. Maschmeyer, Catal. Today, 2011, 178, 164.
19 W. Xu, J. Ni, Q. Zhang, F. Feng, Y. Xiang and X. Li, J. Mater.
Chem. A, 2013, 1, 12811.
20 H. S. Broadbent, L. H. Slaugh and N. L. Jarvis, J. Am. Chem.
Soc., 1954, 76, 1519.
21 L. J. Hoyos, M. Primet and H. Praliaud, J. Chem. Soc., Faraday
Trans., 1992, 88, 3367.
Thermal oxidation was able to recover sulfone deactivated Pd/C
catalyst successfully at temperatures as low as 100 ꢀC. The main
reason for the catalyst deactivation was the blockage of metal
active sites by low valence sulfur species derived from the
hydrogenolysis of 3-nitrophenyl sulfone. The recovery of cata-
lytic activity was attributed to the formation of high valence
sulfur species with the occupation of oxygen under heating,
which largely weakened the strength of Pd–S bond. In this
regard, thermal oxidation method might be highly applicable in
the regeneration of other sulfur-poisoned noble metals.
Furthermore, we found that the regenerated Pd/C could be
recycled for at least 20 times with reasonable loss of activity, so
that thermal oxidation method has good prospects for the
industrialization.
´˜
´
22 S. Ordonez, P. Hurtado and F. V. Dıez, Catal. Lett., 2005, 100,
27.
23 L. Limousy, H. Mahzoul, J. F. Brilhac, F. Garin, G. Maire and
P. Gilot, Appl. Catal., B, 2003, 45, 169.
24 J. M. Jones, V. A. Dupont, R. Brydson, D. J. Fullerton,
N. S. Nasri, A. B. Ross and A. V. K. Westwood, Catal.
Today, 2003, 81, 589.
Acknowledgements
¨
25 C. Schuth, N.-A. Kummer, C. Weidenthaler and H. Schad,
This work was nancially supported by National Natural
Science Foundation of China (NSFC-20976164) and National
Basic Research Program of China (973 Program) (no.
2011CB710800).
Appl. Catal., B, 2004, 52, 197.
26 G. V. Lowry and M. Reinhard, Environ. Sci. Technol., 2000, 34,
3217.
27 M. M. Yung, S. Cheah, K. Magrini-Bair and J. N. Kuhn, ACS
Catal., 2012, 2, 1363.
28 L. Li, C. Howard, D. L. King, M. Gerber, R. Dagle and
D. Stevens, Ind. Eng. Chem. Res., 2010, 49, 10144.
29 S. Cheah, J. L. Olstad, W. S. Jablonski and K. A. Magrini-Bair,
Energy Fuels, 2011, 25, 379.
Notes and references
1 K. Imamura, S.-I. Iwasaki, T. Maeda, K. Hashimoto,
B. Ohtani and H. Kominami, Phys. Chem. Chem. Phys.,
2011, 13, 5114.
30 A. Aguinaga and M. Montes, Appl. Catal., A, 1992, 90, 131.
2 L. Sun, H. Lu and J. Zhou, Dyes Pigm., 2008, 76, 604.
3 T. Zhang, H. Shi, J. Hu and Q. Yang, Trans. Tianjin Univ.,
2010, 16, 161.
¨
31 V. J. G. Kolzer and H.-W. Wassmuth, Ann. Phys., 1986, 498,
472.
¨
32 K. Gotterbarm, N. Luckas, O. Hofert, M. P. A. Lorenz,
4 L. Li, Q. X. Liang, W. Liu, Q. F. Zhang, Q. X. Zhu and X. N. Li,
Fine Chem. Intermed., 2010, 40, 8.
˜
¨
R. Streber, C. Papp, F. Vines, H.-P. Steinruck and
¨
A. Gorling, J. Chem. Phys., 2012, 136, 094702.
5 C. V. Rode, M. J. Vaidya and R. V. Chaudhari, Org. Process Res.
Dev., 1999, 3, 465.
33 W. E. Truce, D. P. Tate and D. N. Burdge, J. Am. Chem. Soc.,
1960, 82, 2872.
6 K. P. Boroujeni, J. Sulfur Chem., 2010, 31, 197.
34 S. Kundu, W. W. Brennessel and W. D. Jones,
Organometallics, 2011, 30, 5147.
¨
7 C. F. J. Konig, T. J. Schildhauer and M. Nachtegaal, J. Catal.,
2013, 305, 92.
35 R. Mozingo, D. E. Wolf, S. A. Harris and K. Folkers, J. Am.
Chem. Soc., 1943, 65, 1013.
36 C. H. Bartholomew, Appl. Catal., A, 2001, 212, 17.
37 F.-B. Zhang and H.-L. Li, Carbon, 2006, 44, 3195.
38 C.-H. Yeh and J.-J. Ho, Chem. Phys. Chem., 2012, 13, 3194.
8 Y. Diao, P. Yang, R. Yin, L. Jiang, L. Wang, H. Zhang, C. Li,
Z. Li and S. Zhang, Appl. Catal., B, 2013, 142–143, 329.
9 S. L. Lakhapatria and M. A. Abraham, Appl. Catal., A, 2011,
405, 149.
10 S. L. Lakhapatri and M. A. Abraham, Appl. Catal., A, 2009,
364, 113.
´
´
´
39 J. C. Rodrıguez, J. Santamarıa and A. Monzon, Appl. Catal., A,
1997, 165, 147.
11 S. L. Lakhapatri and M. A. Abraham, Catal. Sci. Technol.,
2013, 3, 2755.
¨
40 O. Pozdnyakova, D. Teschner, A. Wootsch, J. Krohnert,
B. Steinhauer, H. Sauer, L. Toth, F. C. Jento, A. Knop-
12 G. Ferrari and A. Garberi, US Pat. 3476801, 1969.
13 J. T. Miller and D. C. Koningsberger, J. Catal., 1996, 162, 209.
14 A. M. Venezia, V. La Parola, B. Pawelec and J. L. G. Fierro,
Appl. Catal., A, 2004, 264, 43.
´
¨
Gericke, Z. Paal and R. Schlogl, J. Catal., 2006, 237, 17.
41 L. Peng, A. Philippaerts, X. Ke, J. V. Noyen, F. De Clippel,
G. Van Tendeloo, P. A. Jacobs and B. F. Sels, Catal. Today,
2010, 150, 140.
15 B. Pawelec, R. Mariscal, R. M. Navarro, S. van Bokhorst,
S. Rojas and J. L. G. Fierro, Appl. Catal., A, 2002, 225, 223.
33354 | RSC Adv., 2014, 4, 33347–33354
This journal is © The Royal Society of Chemistry 2014